CN111428627B - A remote sensing extraction method and system for mountainous landforms - Google Patents

Abstract

The invention belongs to the technical field of remote sensing and discloses a method and system for remote sensing extraction of mountain landforms. The remote sensing extraction system for mountain landforms includes an electromagnetic wave detection module, a main control module, a remote sensing image generation module, a correction module, an image segmentation module, a feature extraction module, a remote sensing Data classification module, remote sensing map storage module, display module. The invention effectively solves the segmentation problem caused by the uncertainty of the gray level membership and the uncertainty of the segmentation decision through the image segmentation module, realizes more accurate fitting of the complex histogram distribution characteristics of high-resolution remote sensing data, and It overcomes the noise very well and improves the segmentation accuracy of mountainous landform remote sensing images. At the same time, through the sensing data classification module, using classification system learning and spatio-temporal data self-organization, it can dynamically adjust and improve the hierarchical classification system according to new data, and realize mountainous remote sensing. Dynamic organization and classification management of data.

Description

A remote sensing extraction method and system for mountainous landforms

technical field

The invention belongs to the technical field of remote sensing, and in particular relates to a remote sensing extraction method and system for mountain topography.

Background technique

Remote sensing refers to non-contact, long-distance detection technology. Use sensors/remote sensors to detect the radiation and reflection characteristics of electromagnetic waves of objects. It is a science and technology that obtains the reflected, radiated or scattered electromagnetic wave information (such as electric field, magnetic field, electromagnetic wave, seismic wave, etc.), and extracts, judges, processes, analyzes and applies it. Remote sensing, literally, can be simply understood as remote perception, which generally refers to all non-contact long-distance detection; A new technology that combines monitoring (such as resource management of trees, grasslands, soils, water, minerals, farm crops, fish, and wildlife). However, the segmentation accuracy of the existing mountainous landform remote sensing images is poor; at the same time, the traditional remote sensing data organization is simply stored in the database, and the database is searched and information obtained according to the application requirements. On the one hand, the search efficiency and the ability to obtain information are limited. On the one hand, the information in the database cannot be viewed intuitively, which greatly limits the utilization value of the data.

In summary, the problems in the prior art are:

The segmentation accuracy of the existing remote sensing images of mountainous landforms is poor; at the same time, the traditional remote sensing data organization is simply stored in the database, and the database is searched and information obtained according to the application requirements. On the one hand, the search efficiency and the ability to obtain information are limited. The information in the database cannot be viewed intuitively, which greatly limits the utilization value of the data.

In the prior art, the interference of external factors to electromagnetic waves cannot be effectively removed, and the imaging effect is reduced; the image segmentation effect in the prior art is poor, which is unfavorable to the image processing and analysis effect; in the prior art, the classification quality cannot be guaranteed during data classification processing, and Classification of remote sensing data is slow. In the prior art, the effect of extracting geomorphic features from mountain remote sensing images is poor, and cannot be well applied to terrain surfaces with large undulations.

Contents of the invention

Aiming at the problems existing in the prior art, the invention provides a remote sensing extraction method of mountain topography.

The present invention is achieved in this way, a method for remote sensing extraction of mountainous landforms, the method for remote sensing extraction of mountainous landforms comprises the following steps:

Step 1, by using a remote sensor to detect electromagnetic waves emitted by mountains;

Step 2, using imaging equipment to generate mountain remote sensing images from the detected electromagnetic waves;

Step 3, using correction software to correct the generated mountain remote sensing image; using image segmentation software to segment the mountain remote sensing image;

Step 4: Use image processing software to extract the geomorphological features of mountain remote sensing images; randomly deploy M sensor nodes in the mountain detection site, and perform Delaunay triangulation on the center point of each node; make circumscribed circles of each Delaunay triangle , compare the node radius and the circumcircle radius, if R>r, then there must be a hidden part, save the Delaunay triangle and circumcircle, otherwise remove the circumcircle, the radius of the sensor node is r, and the radius of each circumcircle is R; calculate the remaining The common side length d of two adjacent triangles, if d>2r, or the common side does not intersect with the center line of the circumcircle of the two triangles, then these triangles are clustered and grouped to obtain boundary nodes, and each cluster group will be There is a hidden part; for the center point of the sensor node in each clustering group, use the method of the smallest polygon that can contain the hidden part to show the boundary of the hidden part; determine the false border node for the border node, after removing the false border node, Using the minimum polygon method that can contain hidden parts again, the improved hidden part boundary is shown; the actual coverage area of sensor nodes randomly deployed in mountainous detection parts will become smaller, and the two-dimensional coverage area of sensor nodes after terrain correction is an ellipse. Use the slope and aspect angle to calculate the actual detection radius, use the detection algorithm to calculate the corrected boundary of the hidden part, and finally the topographic features of the remote sensing image of the mountain;

Step five, using data processing software to classify and process the remote sensing data;

In step six, the remote sensing image storage module uses the memory to store the mountain remote sensing image data; and the display module uses the display to display the mountain remote sensing image.

Further, image segmentation methods include:

(1) Read the high-resolution remote sensing image to be segmented;

(2) Using the Gaussian type II fuzzy membership function model of each object category in the high-resolution remote sensing image to be segmented, calculate the Gaussian type II fuzzy membership degree corresponding to each gray level;

(3) Using the segmentation decision-making model of each object category in the high-resolution remote sensing image to be segmented, calculate the membership degree of each gray level in each segmentation decision-making model;

(4) The ground object category corresponding to the maximum membership value of the gray level of each pixel in each segmentation decision model in the high-resolution remote sensing image is the segmentation result;

(5) Change the Gaussian type II fuzzy membership function model according to the set step size and repeat steps (2) to (4), compare all the segmentation results, and take the segmentation result with the highest segmentation accuracy as the final high-resolution remote sensing image segmentation results;

Further, the step (2) includes:

Construct the Gaussian master membership function model and calculate the master degree of membership: perform supervised sampling for each feature category in the high-resolution remote sensing image to be segmented to extract training samples, and calculate the occurrence of each gray level in the training sample in the corresponding feature category The frequency of the Gaussian main membership function model is established for different types of ground objects and the Gaussian main membership degree is calculated;

Determine the uncertainty region of the Gaussian type II fuzzy membership function model: Fuzzify the standard deviation in the Gaussian main membership function model into a standard deviation interval, and the area composed of the Gaussian main membership function model corresponding to the standard deviation interval is Gaussian type II The uncertain region of the fuzzy membership function model, at this time, the Gaussian master membership degree corresponding to each gray level is an interval;

Construct the Gaussian sub-membership function model: determine the mean and variance of the Gaussian sub-membership function model for each gray level within the gray scale range, establish the Gaussian sub-membership function model and calculate the Gaussian sub-membership degree;

Using the Gaussian type II fuzzy membership function model composed of Gaussian main membership function model, Gaussian secondary membership function model and uncertain region, calculate the Gaussian type II fuzzy membership degree: calculate the Gaussian main membership degree of each gray level in the gray scale range The product of the set element and the corresponding Gaussian sub-membership set element is the Gaussian type II fuzzy membership degree of the gray level, and the Gaussian type II fuzzy membership degree corresponding to each gray level is a set;

Further, remote sensing data classification methods include:

1) According to the spatial information and time information of each remote sensing data in the massive remote sensing data, divide the massive remote sensing data into at least one data set, wherein each data set includes at least one remote sensing data;

2) extracting data features in each data set, said data features include: attribute features and image features, attribute features refer to the source, type, resolution of data, image features are histogram features, edge features, texture features;

3) performing hierarchical clustering on the remote sensing data in each data set according to the data characteristics, thereby classifying the remote sensing data with the same data characteristics in each data set into the same data category;

4) Add a semantic label for each data category;

Further, said step 1) includes:

Under the standard ellipsoidal coordinate system, the remote sensing data is coded according to the geographic spatial position to obtain the geographic code of each remote sensing data. The geographic code includes: the data level, longitude and latitude; The height grid is divided and numbered. The geocoding is composed of 20 digits. The first two digits represent the height number, the middle nine digits represent the longitude number, and the last nine digits represent the latitude number;

merging remote sensing data with the same geocoding into the same data set to obtain at least one data set;

In each data set, according to the time information of each remote sensing data, a sequence relationship is established;

Create a spatio-temporal index for data sets that have established a sequence relationship;

Further, in the step 3), the remote sensing data is hierarchically clustered using a hierarchical Chinese restaurant model;

When using the hierarchical Chinese restaurant model to perform hierarchical clustering on any remote sensing data, classify the remote sensing data into an existing data category, or create a new data category, and classify the remote sensing data into the newly established data category;

Use the hierarchical Chinese restaurant model to perform hierarchical clustering on the newly added remote sensing data, classify the newly added remote sensing data into existing data categories, or create a new data category, and classify the newly added remote sensing data into the new Created data categories.

Further, the method of randomly deploying sensor nodes in the mountain detection position, the mountain detection position is expressed as a single-valued function z=h(x, y), the sensing radius r of each sensor is the same, and the sensing area forms a three-dimensional space In is a sphere whose center is the sensor position and r is the radius.

Another object of the present invention is to provide a terminal equipped with a processor for realizing the remote sensing extraction method of mountain landforms.

Another object of the present invention is to provide a computer-readable storage medium, including instructions, which, when run on a computer, cause the computer to execute the method for remote sensing extraction of mountain landforms.

Another object of the present invention is to provide a remote sensing extraction system for mountainous landforms that implements remote sensing extraction of mountainous landforms. The remote sensing extraction system for mountainous landforms includes:

The electromagnetic wave detection module is connected with the main control module, and is used to detect the electromagnetic wave emitted by the mountain through the remote sensor;

The main control module is connected with the electromagnetic wave detection module, the remote sensing image generation module, the correction module, the image segmentation module, the feature extraction module, the remote sensing data classification module, the remote sensing image storage module and the display module, and is used to control the normal operation of each module through the single-chip microcomputer;

The remote sensing image generation module is connected with the main control module, and is used to generate the mountain remote sensing image from the detected electromagnetic wave through the imaging device;

The correction module is connected with the main control module, and is used for correcting the generated mountain remote sensing image through correction software;

The image segmentation module is connected with the main control module, and is used for segmenting the mountain remote sensing image through the image segmentation software;

The feature extraction module is connected with the main control module, and is used to extract the geomorphic features of the remote sensing images of mountains through image processing software;

The remote sensing data classification module is connected with the main control module, and is used to classify and process the remote sensing data through data processing software;

The remote sensing image storage module is connected with the main control module and is used to store mountain remote sensing image data through memory;

The display module is connected with the main control module and is used for displaying mountain remote sensing images through the display.

Advantage of the present invention and positive effect are:

The invention constructs a Gaussian type II fuzzy membership function model for the image through the image segmentation module and constructs a segmentation decision model by weighting all membership degrees, effectively solving the segmentation problems caused by the uncertainty of gray level membership and the uncertainty of segmentation decision problem, to achieve a more accurate fitting of the complex histogram distribution characteristics of high-resolution remote sensing data, and to overcome the noise well, improving the segmentation accuracy of mountainous landform remote sensing images; at the same time, through the sensory data classification module, the classification system is used Spatial-temporal data self-organization, automatically establishes a hierarchical classification system based on the data, and provides hierarchically refined clustering results from coarse to fine, flexible and convenient, and can dynamically adjust and improve the hierarchical classification system based on new data in the subsequent use process , realize the dynamic organization and classification management of mountain remote sensing data.

The imaging device of the present invention uses the PURE-LET wavelet domain denoising model in the process of generating mountain remote sensing images to effectively remove the interference of external factors on electromagnetic waves and improve the imaging effect; the computer utilizes image segmentation software and uses FCM image segmentation algorithms to effectively improve Segmentation effect of mountainous remote sensing images, speed up the segmentation speed, increase the segmentation speed, and improve the image processing and analysis effect; use data processing software to classify remote sensing data using Naive Bayesian algorithm, and improve the accuracy of remote sensing data under the condition of ensuring the quality of classification classification speed.

In the present invention, M sensor nodes are randomly deployed in the mountain detection site, and the center point of each node is divided into Delaunay triangles; the circumscribed circle of each Delaunay triangle is made, and the radius of the node is compared with the radius of the circumscribed circle, if R>r , then there must be a hidden part, save the Delaunay triangle and circumcircle, otherwise remove the circumcircle, the radius of the sensor node is r, and the radius of each circumcircle is R; calculate the common side length d of the remaining two adjacent triangles, if d >2r, or the common side does not intersect with the center line of the circumcircle of the two triangles, then these triangles are clustered and grouped to obtain boundary nodes, and each cluster group will have a hidden part; for each cluster group The sensor The center point of the node uses the method of the smallest polygon that can contain the hidden part to show the boundary of the hidden part; the boundary node is judged as a false border node, and after the false border node is removed, the method of the smallest polygon that can include the hidden part is used to express the boundary of the hidden part The improved boundary of the concealed part; the actual coverage area of the sensor nodes randomly deployed in the mountain detection site will become smaller, and the two-dimensional coverage area of the sensor nodes after terrain correction is an ellipse, and the actual detection radius is calculated by using the slope and aspect angle, and the detection is used The algorithm calculates the corrected boundary of the concealed part, and finally the geomorphic features of the mountain remote sensing image; it can be well used on the terrain surface with large undulations.

Description of drawings

Fig. 1 is a flow chart of a remote sensing extraction method for mountain topography provided by an embodiment of the present invention.

Fig. 2 is a structural diagram of a remote sensing extraction system for mountain topography provided by an embodiment of the present invention.

In the figure: 1. Electromagnetic wave detection module; 2. Main control module; 3. Remote sensing image generation module; 4. Calibration module; 5. Image segmentation module; 6. Feature extraction module; 7. Remote sensing data classification module; 8. Remote sensing map storage module; 9. display module.

Detailed ways

In order to further understand the invention content, features and effects of the present invention, the following embodiments are exemplified, and detailed descriptions are included with the accompanying drawings.

The structure of the present invention will be described in detail below in conjunction with the accompanying drawings.

As shown in Figure 1, the remote sensing extraction method of mountain landform provided by the present invention comprises the following steps:

S101, using remote sensors to detect and collect electromagnetic waves emitted from mountains.

S102, using an imaging device with a PURE-LET wavelet domain denoising model to generate a mountain remote sensing image from the detected electromagnetic waves.

S103, correcting the generated mountain remote sensing image by using the correction software through the computer. The remote sensing images of mountains are segmented by computer using image segmentation software using FCM image segmentation algorithm.

S104, using image processing software to extract geomorphic features of the mountain remote sensing image through a computer. The computer uses data processing software to classify and process the remote sensing data using the Naive Bayesian algorithm.

S105, storing mountain remote sensing image data in memory. And display mountain remote sensing images through the monitor.

In step S102, the imaging device provided by the embodiment of the present invention uses the PURE-LET wavelet domain denoising model in the process of generating mountain remote sensing images to effectively remove the interference of external factors on electromagnetic waves and improve the imaging effect. The specific algorithm is:

均写成一组基本阈值函数的线性组合：At each scale the wavelet coefficients are estimated

Both are written as a linear combination of a set of basic threshold functions:

And determine the coefficient vector a=[a ₁ , . . . , a _M ] ^T through the minimization of PURE.

Let θ(d, s) = θ ^j (d ^j , s ^j ) be an estimate of the noise-free wavelet coefficient δ = δ ^j .

The functions θ ⁺ (d, s) and θ ^- (d, s) include:

为/>

的标准基，除e_k(k)＝1外其余元素均为0。则随机变量PURE_j为子带j下MSE的无偏估计，即E{PURE_j}＝E{MSE_j}。in,

for />

The standard base of , all elements are 0 except e _k (k)=1. Then the random variable PURE _j is an unbiased estimate of the MSE under subband j, that is, E{PURE _j }=E{MSE _j }.

Through the minimization of PURE, the linear combination parameters estimated by the wavelet in formula (2) are calculated. Substituting formula (2) into formula (3), and omitting the independent variable (d, s), we have

In step S103, the FCM image segmentation algorithm provided by the embodiment of the present invention can effectively improve the segmentation effect of mountainous remote sensing images, speed up the segmentation speed, and improve the image processing and analysis effect by using the image segmentation software provided by the computer. Specific steps:

(1) Determination of initialization: According to the requirements of image segmentation, it is necessary to determine the initialization of the image, initialize the required parameters, and set the clustering center of the histogram.

(2) The determination of the adaptability of the factor, the degree of fitness, according to the constructed fitness function:

f=a/(b+J).

Among them, a and b are adjustable parameters, which can be set to 10 and 1.5 respectively according to the experiment, and J is the objective function.

(3) Mutation operation: the amount of change before and after the individual is 0.5r(t/T), the data r is a random number generated within a specified interval, and T is the maximum number of calculations.

(4) Iterative calculation: the new fuzzy membership degree matrix will be obtained through the new cutting data, and new cutting parameters will be generated. Return to step 2 for iterative calculation until the condition is terminated and the image segmentation is completed.

In step S104, the data processing software provided by the embodiment of the present invention adopts the naive Bayesian algorithm to classify the remote sensing data, and under the condition of ensuring the classification quality, improve the classification speed of the remote sensing data. The specific algorithm is:

Let D be the set of class labels associated with training objects. Each object is represented by an n-dimensional attribute vector X={x ₁ , x ₂ ... x _n }, which describes the values of n attributes A ₁ , A ₂ ... A _n . Assuming that the original set is divided into m classes C ₁ , C ₂ ... C _m based on n-dimensional attributes, calculate the posterior probability of each class pair X, and assign the object X to the class with the highest posterior probability. The formula for calculating the posterior probability P(C _i |X) is:

Due to the large calculation overhead of P(C _i |X), the conditional independence of classes is assumed, the class label of the vector is given, and the attribute values are assumed to be conditionally independent of each other. The calculation formula of P(X _i |C) is:

Among them, P(x ₁ |C _i )P(x ₂ |C _i )...P(x _n |C _n ) can be easily calculated by the training object, and x _k represents the value of X on attribute A _k . Compute P(X|C _i )P(C _i ) for each class C _i . When P(X|C _i )P(C _i )>P(X|C _j )P(C _j ), 1≤j≤m, j≠i holds, X belongs to class C _i .

In step S104, M sensor nodes are randomly deployed in the mountain detection site, and Delaunay triangulation is performed on the center point of each node.

Make the circumcircle of each Delaunay triangle, compare the node radius and the circumcircle radius, if R>r, then there must be a hidden part, save the Delaunay triangle and circumcircle, otherwise remove the circumcircle, the sensor node radius is r, each The radius of the circumcircle is R; calculate the length d of the common side of the remaining two adjacent triangles, if d>2r, or the common side does not intersect the center line of the circumcircle of the two triangles, then cluster and group these triangles to get Boundary nodes, each clustering group will have a hidden part; for the center point of the sensor node in each clustering group, use the method of the smallest polygon that can contain the hidden part to indicate the boundary of the hidden part; make a false boundary node for the boundary node After removing the false boundary nodes, the minimum polygon method that can contain the hidden part is used again to show the improved hidden part boundary; the actual coverage area of the sensor nodes randomly deployed in the mountain detection part will become smaller, and the sensor after terrain correction The two-dimensional coverage area of the node is an ellipse, and the actual detection radius is calculated by using the slope and aspect angle, and the corrected boundary of the hidden part is calculated by using the detection algorithm, and finally the topographic features of the mountain remote sensing image.

The method of randomly deploying sensor nodes in the mountain detection position, the mountain detection position is expressed as a single-valued function z=h(x, y), the sensing radius r of each sensor is the same, and the sensing area forms a three-dimensional space as A sphere with the sensor position as the center and r as the radius.

The calculation method of the actual detection radius is on the curved surface z=h(x, y), and the direction gradient for point P(x, y) is:

和/>

分别为x和y方向的偏导数，i和j为单位矢量，方向梯度的模为坡度；in

and />

are the partial derivatives in the x and y directions, respectively, i and j are unit vectors, and the modulus of the directional gradient is the slope;

The slope G of point P along the β direction is:

G=Scosβ

β is the slope aspect. Due to the undulating defects of the three-dimensional terrain, the relationship between the actual detection radius r’ of the sensor node along the β direction and the ideal detection radius r is expressed as:

r'=rcosγ

The relationship between the actual detection radius r' and the slope S and aspect angle β is:

r'=rcos(arctan(Scosβ)).

The correction method is along the slope direction, the difference between the two contour lines intersected by the node is the height difference Δh, the distance between the two intersecting contour lines is Δd, and the slope S is expressed as:

Calculate the ellipse projection of each sensor node on the two-dimensional plane under the three-dimensional terrain.

As shown in Figure 2, the remote sensing extraction system for mountain landforms provided by the embodiment of the present invention includes:

Electromagnetic wave detection module 1, main control module 2, remote sensing image generation module 3, correction module 4, image segmentation module 5, feature extraction module 6, remote sensing data classification module 7, remote sensing image storage module 8, display module 9.

The electromagnetic wave detection module 1 is connected with the main control module 2, and is used for detecting electromagnetic waves emitted by mountains through remote sensors.

The main control module 2 is connected with the electromagnetic wave detection module 1, the remote sensing image generation module 3, the correction module 4, the image segmentation module 5, the feature extraction module 6, the remote sensing data classification module 7, the remote sensing map storage module 8, and the display module 9, for The normal operation of each module is controlled by the single chip microcomputer.

The remote sensing image generating module 3 is connected with the main control module 2, and is used to generate a mountain remote sensing image from the detected electromagnetic waves through the imaging device.

The correction module 4 is connected with the main control module 2, and is used for correcting the generated mountain remote sensing image through correction software.

The image segmentation module 5 is connected with the main control module 2, and is used for segmenting the remote sensing images of mountains through image segmentation software.

The feature extraction module 6 is connected with the main control module 2, and is used for extracting the geomorphological features of the mountain remote sensing image through image processing software.

The remote sensing data classification module 7 is connected with the main control module 2, and is used for classifying and processing the remote sensing data through data processing software.

The remote sensing image storage module 8 is connected with the main control module 2 and is used for storing mountain remote sensing image data through a memory.

The display module 9 is connected with the main control module 2, and is used for displaying mountain remote sensing images through a display.

Image segmentation module 5 segmentation methods provided by the invention include:

(1) Read the high-resolution remote sensing image to be segmented.

(2) Using the Gaussian type II fuzzy membership function model of each object category in the high-resolution remote sensing image to be segmented, calculate the Gaussian type II fuzzy membership degree corresponding to each gray level.

(3) Using the segmentation decision-making model of each object category in the high-resolution remote sensing image to be segmented, calculate the membership degree of each gray level in each segmentation decision-making model.

(4) The ground object category corresponding to the maximum membership value of the gray level of each pixel in each segmentation decision model in the high-resolution remote sensing image is the segmentation result.

(5) Change the Gaussian type II fuzzy membership function model according to the set step size and repeat steps (2) to (4), compare all the segmentation results, and take the segmentation result with the highest segmentation accuracy as the final high-resolution remote sensing image Split results.

Step (2) provided by the present invention comprises:

Construct the Gaussian master membership function model and calculate the master degree of membership: perform supervised sampling for each feature category in the high-resolution remote sensing image to be segmented to extract training samples, and calculate the occurrence of each gray level in the training sample in the corresponding feature category The frequency of the Gaussian main membership function model is established for different surface object categories and the Gaussian main membership degree is calculated.

Determine the uncertainty region of the Gaussian type II fuzzy membership function model: Fuzzify the standard deviation in the Gaussian main membership function model into a standard deviation interval, and the area composed of the Gaussian main membership function model corresponding to the standard deviation interval is Gaussian type II The uncertain region of the fuzzy membership function model, at this time, the Gaussian master membership degree corresponding to each gray level is an interval.

Construct the Gaussian sub-membership function model: determine the mean and variance of the Gaussian sub-membership function model for each gray level within the gray scale range, establish the Gaussian sub-membership function model and calculate the Gaussian sub-membership degree.

Using the Gaussian type II fuzzy membership function model composed of Gaussian main membership function model, Gaussian secondary membership function model and uncertain region, calculate the Gaussian type II fuzzy membership degree: calculate the Gaussian main membership degree of each gray level in the gray scale range The product of the set element and the corresponding Gaussian sub-membership set element is the Gaussian type II fuzzy membership degree of the gray level, and the Gaussian type II fuzzy membership degree corresponding to each gray level is a set.

Remote sensing data classification module 7 classification methods provided by the present invention include:

1) According to the spatial information and time information of each remote sensing data in the massive remote sensing data, divide the massive remote sensing data into at least one data set, wherein each data set includes at least one remote sensing data.

2) Extract the data features in each data set, the data features include: attribute features and image features, the attribute features refer to the source, type, resolution of the data, and the image features are histogram features, edge features, and texture features.

3) According to the data characteristics, perform hierarchical clustering on the remote sensing data in each data set, so as to classify the remote sensing data with the same data characteristics in each data set into the same data category.

4) Add a semantic label to each data category.

Step 1) provided by the present invention comprises:

Under the standard ellipsoidal coordinate system, the geographical space position of each remote sensing data is coded to obtain the geographical code of each remote sensing data, and the geographical code includes: data level, longitude and latitude. The global spatial range is divided and numbered according to latitude, longitude, and height grids. The geocoding consists of 20 digits, the first two digits represent the height number, the middle nine digits represent the longitude number, and the last nine digits represent the latitude number.

The remote sensing data with the same geocoding are merged into the same data set to obtain at least one data set.

In each data set, according to the time information of each remote sensing data, a sequence relationship is established.

Create a spatio-temporal index for data sets that have established a sequence relationship.

In step 3) provided by the present invention, hierarchical clustering is performed on the remote sensing data using a hierarchical Chinese restaurant model.

When using the hierarchical Chinese restaurant model to perform hierarchical clustering on any remote sensing data, the remote sensing data is classified into an existing data category, or a new data category is created, and the remote sensing data is classified into the newly established data category.

Use the hierarchical Chinese restaurant model to perform hierarchical clustering on the newly added remote sensing data, classify the newly added remote sensing data into existing data categories, or create a new data category, and classify the newly added remote sensing data into the new Created data classes.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention in any form. Any simple modifications made to the above embodiments according to the technical essence of the present invention, equivalent changes and modifications, all belong to this invention. within the scope of the technical solution of the invention.

Claims (10)

1. A mountain landform remote sensing extraction method is characterized by comprising the following steps:

firstly, detecting electromagnetic waves emitted from a mountain land by using a remote sensor;

secondly, generating a mountain remote sensing image by using the detected electromagnetic waves through an imaging device;

correcting the generated mountain remote sensing image by using correction software; segmenting the mountain remote sensing image by using image segmentation software;

extracting the landform characteristics of the mountain remote sensing image by using image processing software; randomly deploying M sensor nodes in a mountain detection part, and performing Delaunay triangle subdivision on the central point of each node; making a circumscribed circle of each Delaunay triangle, comparing the radius of the node with the radius of the circumscribed circle, if R is greater than R, determining that a hidden part exists, storing the Delaunay triangle and the circumscribed circle, and if not, removing the circumscribed circle, wherein the radius of the node of the sensor is R, and the radius of each circumscribed circle is R; calculating the public side length d of the rest two adjacent triangles, and if d is greater than 2r or the public side is not intersected with the central connecting line of the circumscribed circles of the two triangles, clustering and grouping the triangles to obtain boundary nodes, wherein each clustering and grouping can have a hidden part; for the center point of the sensor node in each cluster group, representing the boundary of the hidden part by using a method of a minimum polygon capable of containing the hidden part; judging a false boundary node for the boundary node, after removing the false boundary node, expressing the improved boundary of the hidden part by using a minimum polygon method capable of containing the hidden part again; the actual coverage area of the sensor nodes randomly deployed at the mountain land detection part is reduced, the two-dimensional coverage area of the sensor nodes subjected to terrain correction is an ellipse, the actual detection radius is calculated by utilizing the slope and the slope angle, the corrected hidden part boundary is calculated by utilizing a detection algorithm, and finally the landform characteristics of the mountain land remote sensing image are obtained;

step five, carrying out classification processing operation on the remote sensing data by using data processing software;

step six, storing the mountain remote sensing image data by using a memory through a remote sensing image storage module; and the mountain remote sensing image is displayed by the display module through the display.

2. The remote sensing extraction method of mountain land features as claimed in claim 1, wherein the image segmentation method comprises:

(1) Reading a high-resolution remote sensing image to be segmented;

(2) Calculating the Gaussian two-type fuzzy membership degree corresponding to each gray level by using a Gaussian two-type fuzzy membership function model of each feature class in the high-resolution remote sensing image to be segmented;

(3) Calculating the membership degree of each gray level in each segmentation decision model by using the segmentation decision model of each object class in the high-resolution remote sensing image to be segmented;

(4) The ground object category corresponding to the maximum membership value of the gray level of each pixel in each segmentation decision model in the high-resolution remote sensing image is a segmentation result;

(5) And (3) changing the Gaussian two-type fuzzy membership function model according to the set step length, repeating the steps (2) to (4), comparing all segmentation results, and taking the segmentation result with the highest segmentation precision as a final high-resolution remote sensing image segmentation result.

3. The remote sensing extraction method of mountain land features as claimed in claim 2, wherein said step (2) comprises:

constructing a Gaussian main membership function model and calculating a main membership degree: carrying out supervised sampling on each ground feature class in the high-resolution remote sensing image to be segmented to extract a training sample, calculating the frequency of each gray level in the training sample appearing in the corresponding ground feature class, establishing a Gaussian main membership function model for different ground feature classes and calculating the Gaussian main membership degree;

determining an uncertain region of the Gaussian two-type fuzzy membership function model: the standard difference in the Gaussian main membership function model is fuzzified into a standard difference interval, a region formed by the Gaussian main membership function model corresponding to the standard difference interval is an uncertain region of the Gaussian two-type fuzzy membership function model, and at the moment, the Gaussian main membership degree corresponding to each gray level is an interval;

constructing a Gaussian subordination function model: determining the mean value and the variance of a Gaussian subordination function model of each gray level in the gray level range, establishing a Gaussian subordination function model and calculating the Gaussian subordination degree;

calculating the Gaussian second type fuzzy membership degree by utilizing a Gaussian second type fuzzy membership function model consisting of a Gaussian main membership function model, a Gaussian secondary membership function model and an uncertain region: and calculating the product of the Gaussian primary membership set element and the corresponding Gaussian secondary membership set element of each gray level in the gray level range, namely the Gaussian secondary fuzzy membership of the gray level, wherein the Gaussian secondary fuzzy membership corresponding to each gray level is a set.

4. The mountain landform remote sensing extraction method as claimed in claim 1, wherein the remote sensing data classification method comprises:

1) Dividing the mass remote sensing data into at least one data set according to the spatial information and the time information of each remote sensing data in the mass remote sensing data, wherein each data set comprises at least one remote sensing data;

2) Extracting data features in each data set, the data features comprising: the system comprises attribute features and image features, wherein the attribute features refer to the source, type and resolution of data, and the image features refer to histogram features, edge features and texture features;

3) According to the data characteristics, carrying out hierarchical clustering on the remote sensing data in each data set, so as to classify the remote sensing data with the same data characteristics in each data set into the same data category;

4) A semantic tag is added for each data category.

5. The remote sensing extraction method of mountain landform as claimed in claim 4, wherein the step 1) comprises:

coding each remote sensing data according to a geographic space position under a standard ellipsoid coordinate system to obtain a geocode of each remote sensing data, wherein the geocode comprises: the level, longitude and latitude of the data; dividing a global space range according to longitude, latitude and altitude grids and numbering, wherein the geocode consists of 20 bits, the first two bits represent altitude numbers, the middle 9 bits represent longitude numbers, and the last 9 bits represent latitude numbers;

merging the remote sensing data with the same geocode into the same data set to obtain at least one data set;

in each data set, establishing a sequence relation according to the time information of each remote sensing data;

and establishing a space-time index for the data set with the established sequence relation.

6. The remote sensing extraction method of mountain landform as claimed in claim 4, wherein in step 3), a hierarchical restaurant model is used to perform hierarchical clustering on the remote sensing data;

when a restaurant model in a hierarchy is adopted to carry out hierarchical clustering on any remote sensing data, classifying the remote sensing data into an existing data category, or newly establishing a data category, and classifying the remote sensing data into the newly established data category;

and carrying out hierarchical clustering on the newly-added remote sensing data by adopting a restaurant-in-hierarchy model, classifying the newly-added remote sensing data into an existing data category, or newly establishing a data category, and classifying the newly-added remote sensing data into the newly-established data category.

7. The remote sensing extraction method of mountain land features as claimed in claim 1, wherein the sensor nodes are randomly deployed in mountain land detection positions, the mountain land detection positions are expressed as a single-valued function z = h (x, y), the sensing radius r of each sensor is the same, and the sensing area forms a sphere centered on the sensor position and having a radius r in a three-dimensional space.

8. A terminal, characterized in that the terminal is equipped with a processor for implementing the mountain landform remote sensing extraction method of any one of claims 1 to 7.

9. A computer-readable storage medium comprising instructions that, when executed on a computer, cause the computer to perform the remote mountain relief feature extraction method of any one of claims 1-7.

10. A mountain land feature remote sensing extraction system for implementing the mountain land feature remote sensing extraction method of claim 1, wherein the mountain land feature remote sensing extraction system comprises:

the electromagnetic wave detection module is connected with the main control module and used for detecting the electromagnetic waves emitted by the mountainous region through a remote sensor;

the main control module is connected with the electromagnetic wave detection module, the remote sensing image generation module, the correction module, the image segmentation module, the feature extraction module, the remote sensing data classification module, the remote sensing image storage module and the display module and is used for controlling each module to normally work through the single chip microcomputer;

the remote sensing image generation module is connected with the main control module and used for generating a mountain remote sensing image from the detected electromagnetic waves through the imaging equipment;

the correction module is connected with the main control module and used for correcting the generated mountain remote sensing image through correction software;

the image segmentation module is connected with the main control module and is used for carrying out segmentation operation on the mountain remote sensing image through image segmentation software;

the characteristic extraction module is connected with the main control module and used for extracting the landform characteristics of the mountain remote sensing image through image processing software;

the remote sensing data classification module is connected with the main control module and is used for performing classification processing operation on the remote sensing data through data processing software;

the remote sensing image storage module is connected with the main control module and used for storing mountain remote sensing image data through the memory;

and the display module is connected with the main control module and used for displaying the mountain remote sensing image through the display.

Images